[SPIKE] Distributed logging and traceability

[loripusey]

User Story

As an engineer, I would like to easily trace the logs across all the services, so that I can interpret the workflow by looking at the logs

Tasks

• Determine how we might implement this

Description

Currently, veteran facing side team uses lorotaUuid to correlate the logs across vets-api, CHIP, Lorota & Clinician workflow. As LoROTA generates temporary lorotaUUID for day-of check-in & pre-check-in, downstream service Vista doesn’t have knowledge on lorotaUUID & developers get lost beyond LoROTA while debugging any failure.

One approach could be to define a request-id in the header while initiating any request & pass it across all the services to include in logs. It enables not only identifying logs in Vista for any user workflow but will also help in segregating logs by workflow origination.

For example, consider a scenario where a veteran tries to check in but is dropped with a failure in the system & a staff member checking in the veteran from the staff facing app. Currently, when the issue gets reported to the developer, debug logs show only halfway through the workflow with lorotaUUID & engineers cannot identify it from debug log until verbally told about the staff member initiated check-in.

[loripusey]

@kanchanasuriya Do you know what this ticket is for?

[kanchanasuriya]

@loripusey Updated the description, pls find the details.

[loripusey]

too old, closing